Universal solvation model based on conductor-like screening model

Citation
Dm. Dolney et al., Universal solvation model based on conductor-like screening model, J COMPUT CH, 21(5), 2000, pp. 340-366
Citations number
45
Categorie Soggetti
Chemistry
Journal title
JOURNAL OF COMPUTATIONAL CHEMISTRY
ISSN journal
01928651 → ACNP
Volume
21
Issue
5
Year of publication
2000
Pages
340 - 366
Database
ISI
SICI code
0192-8651(20000415)21:5<340:USMBOC>2.0.ZU;2-B
Abstract
Atomic surface tensions are parameterized for use with salvation models in which the electrostatic part of the calculation is based on the conductor-l ike screening model (COSMO) and the semiempirical molecular orbital methods AM1, PM3, and MNDO/d. The convergence of the calculated polarization free energies with respect to the numerical parameters of the electrostatic calc ulations is first examined. The accuracy and precision of the calculated va lues are improved significantly by adjusting two parameters that control th e segmentation of the solvent-accessible surface that is used for the calcu lations. The accuracy of COSMO calculations is further improved by adopting an optimized set of empirical electrostatic atomic radii. Finally, the ele ctrostatic calculation is combined with SM5-type atomic surface tension fun ctionals that are used to compute the nonelectrostatic portions of the solv ation free energy. All parameterizations are carried out using rigid (R) ga s-phase geometries; this combination (SM5-type surface tensions, COSMO elec trostatics, and rigid geometries) is called SM5CR. Six air-water and 76 wat er-solvent partition coefficients are added to the training set of air-solv ent data points previously used to parameterize the SM5 suite of solvation models, thereby bringing the total number of data points in the training se t to 2266. The model yields free energies of solvation and transfer with me an unsigned errors of 0.63, 0.59, and 0.61 kcal/mol for AM1, PM3, and MNDO/ d, respectively, over all 2217 data points for neutral solutes in the train ing set and mean unsigned errors of 3.0, 2.7, and 3.1 kcal/mol, respectivel y, for 49 data points for the ions. (C) 2000 John Wiley & Sons, Inc.